The unique transmission of the Y chromosome provides an unusual opportunity to examine evolutionary forces at several levels of biological organization. The theoretical and experimental approaches outlined in this proposal represent an attempt to explore the population genetics of variation in the Drosophila Y chromosome, with the intention of gaining insight into sex chromosome evolution through understanding the significance of extant variation in the Y chromosome. The theoretical study is divided into three sections. The first is the development and analysis of "classical" selection models, where discrete variation in Y chromosomes is assumed, and systems of equations will be derived that predict changes in allele frequencies under a variety of conditions, including meiotic drive, X-Y recombination, viability effects, fertility selection, sexual selection, mutation and migration. The second theoretical problem is to develop statistical analyses of population samples in the spirit of Christiansen and Frydenburg's "selection component analysis". Finally, an analysis of the infinite alleles model for the Y chromosome will be done, and the utility of goodness-of-fit tests (including Ewens-Watterson sampling theory and ad hoc methods) for testing the neutral model will be examined. Three experimental projects will coincide with the theory that is developed. A series of Y chromosome replacement lines has been constructed and progeny segregation suggests a slight but significant variation among Y chromosomes. Additional experiments will examine the possibility of Y-mediated variation in male fertility. A survey of natural populations of Drosophila melanogaster using Southern analysis with Y-homologous probes will be undertaken. Data from this study will be analyzed to ascertain goodness-of-fit to neutral models, and to explore population structuring of this male-limited genetic element. Drosophila will be collected from a local population in order to perform a selection component analysis. This will provide a powerful statistical test of the significance of Y chromosome variation in a natural population. The integrated approach with a close interplay between theoretical and experimental methods should prove particularly fruitful in this study.